Constituency Parsing using LLMs
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909807201484800 |
|---|---|
| author | Bai, Xuefeng Wu, Jialong Chen, Yulong Wang, Zhongqing Chen, Kehai Zhang, Min Zhang, Yue |
| author_facet | Bai, Xuefeng Wu, Jialong Chen, Yulong Wang, Zhongqing Chen, Kehai Zhang, Min Zhang, Yue |
| contents | Constituency parsing is a fundamental yet unsolved challenge in natural language processing. In this paper, we examine the potential of recent large language models (LLMs) to address this challenge. We reformat constituency parsing as a sequence-to-sequence generation problem and evaluate the performance of a diverse range of LLMs under zero-shot, few-shot, and supervised fine-tuning learning paradigms. We observe that while LLMs achieve acceptable improvements, they still encounter substantial limitations, due to the absence of mechanisms to guarantee the validity and faithfulness of the generated constituent trees. Motivated by this observation, we propose two strategies to guide LLMs to generate more accurate constituent trees by learning from erroneous samples and refining outputs in a multi-agent collaboration way, respectively. The experimental results demonstrate that our methods effectively reduce the occurrence of invalid and unfaithful trees, thereby enhancing overall parsing performance and achieving promising results across different learning paradigms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_19462 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Constituency Parsing using LLMs Bai, Xuefeng Wu, Jialong Chen, Yulong Wang, Zhongqing Chen, Kehai Zhang, Min Zhang, Yue Computation and Language Constituency parsing is a fundamental yet unsolved challenge in natural language processing. In this paper, we examine the potential of recent large language models (LLMs) to address this challenge. We reformat constituency parsing as a sequence-to-sequence generation problem and evaluate the performance of a diverse range of LLMs under zero-shot, few-shot, and supervised fine-tuning learning paradigms. We observe that while LLMs achieve acceptable improvements, they still encounter substantial limitations, due to the absence of mechanisms to guarantee the validity and faithfulness of the generated constituent trees. Motivated by this observation, we propose two strategies to guide LLMs to generate more accurate constituent trees by learning from erroneous samples and refining outputs in a multi-agent collaboration way, respectively. The experimental results demonstrate that our methods effectively reduce the occurrence of invalid and unfaithful trees, thereby enhancing overall parsing performance and achieving promising results across different learning paradigms. |
| title | Constituency Parsing using LLMs |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2310.19462 |